Impact of alkoxy chain length on carbazole-based, visible light-driven, dye sensitized photocatalytic hydrogen production

Article abstract of DOI:10.1039/c5ta04991a

Alkoxyphenyl-substituted carbazole-based metal-free organic dyes were synthesized and effectively used for dye-sensitized, visible-light-driven, photocatalytic hydrogen production. Photocatalytic hydrogen production was investigated using a TiO₂/dye/Pt structure with triethanolamine as the sacrificial reagent. The dye-loaded TiO₂ photocatalyst exhibited a high yield of hydrogen production when the length of the alkoxy chain was long enough to sufficiently improve the hydrophobicity at the interface between the dye-loaded TiO₂ and the water medium. In the alkoxyphenyl-substituted carbazole dyes, the dye with the longest alkoxy chain (C22) exhibited the best hydrogen production performance, but it had a yield only slightly better than that of the dye with the second longest chain length (C16). The dye C22 displayed a turnover number (TON) of 3094 after 24 h of visible light irradiation (>420 nm). However, the compound with no hydrophobic substituent (C1), exhibited the lowest hydrogen production performance with a TON of 1497. Thus, a 207% increase in the hydrogen production yield was observed when hydrophobic substituents were present. Analysis of time-resolved absorption spectra, impedance spectra and incident photon conversion efficiency spectra revealed that the alkoxy chain has a hydrophobic effect at the interface between the dye-loaded TiO₂ and the water. Specifically, the hydrophobicity of the dye improved the charge-recombination lifetime for electron injection from the dye into the TiO₂ surface in the water for hydrogen production.

Full text of DOI:10.1039/c5ta04991a

View Article Online
View Journal
Journal of
Materials Chemistry A
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: M. Watanabe, H.
Hagiwara, Y. Ogata, A. Staykov, S. Bishop, N. H. Perry, Y. J. Chang, S. Ida, K. Tanaka and T. Ishihara, J.
Mater. Chem. A, 2015, DOI: 10.1039/C5TA04991A.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/materialsA

Page 1 of 9
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C5TA04991A
Journal of Materials Chemistry A
RSCPublishing
PAPER
Impact of Alkoxy Chain Length on Carbazole-
based, Visible Light-driven, Dye Sensitized
Photocatalytic Hydrogen Production
Cite this: DOI: 10.1039/x0xx00000x
Motonori Watanabe*a, Hidehisa Hagiwaraa,b, Yudai Ogatab, Aleksandar
Staykova, Sean R. Bishopa, Nicola H. Perrya, Yuan Jay Chang,c Shintaro Idaa,b,
Keiji Tanakaa,b, Tatsumi Ishihara*a,b
Received 00th January 2012,
Accepted 00th January 2012
DOI: 10.1039/x0xx00000x
www.rsc.org/
Alkoxyphenylꢀsubstituted carbazoleꢀbased metalꢀfree organic dyes were synthesized and
effectively used for dyeꢀsensitized, visibleꢀlightꢀdriven, photocatalytic hydrogen production.
Photocatalytic hydrogen production was investigated using a TiO2/dye/Pt structure with
triethanolamine as the sacrificial reagent. The dyeꢀloaded TiO2 photocatalyst exhibited a high
yield of hydrogen production when the length of the alkoxy chain was long enough to
sufficiently improve the hydrophobicity at the interface between the dyeꢀloaded TiO2 and the
water medium. In the alkoxyphenylꢀsubstituted carbazole dyes, the dye with the longest alkoxy
chain (C22) exhibited the best hydrogen production performance, but it had a yield only slightly
better than that of the dye with the second longest chain length (C16). The dye C22 displayed a
turnover number (TON) of 3094 after 24 h of visible light irradiation (>420 nm). However, the
compound with no hydrophobic substituent (C1), exhibited the lowest hydrogen production
performance with a TON of 1497. Thus, a 207% increase in the hydrogen production yield was
observed when hydrophobic substituents were present. Analysis of timeꢀresolved absorption
spectra, impedance spectra and incident photon conversion efficiency spectra revealed that the
alkoxy chain has a hydrophobic effect at the interface between the dyeꢀloaded TiO2 and the
water. Specifically, the hydrophobicity of the dye improved the chargeꢀrecombination lifetime
for electron injection from the dye into the TiO2 surface in the water for hydrogen production.
An alternative approach involves the use of dyeꢀsensitized
photocatalysts. This method may cover the visible range, because the
molecular structures of organic dyes can be readily modified. 6,7 In a
Introduction
Production of hydrogen is key for establishing clean energy
systems because it can be used as a clean energy source in hydrogen
fuel cells.1,2 Hydrogen production using solar lightꢀbased catalysts is
an ideal generation method because it is a simple and clean reaction:
i) only a photocatalyst, sunlight and water are required; ii) the
reaction occurs under mild conditions (even room temperature) and
iii) water is split into hydrogen without producing carbon dioxide as
a byproduct. Although this method has many environmental
advantages, many challenges must be overcome before
photocatalytic water splitting can be practically implemented. Most
current photocatalysts do not exhibit good water splitting activity
when exposed to visible light and require ultraviolet (UV) light
because of their large energy band gaps. A photocatalyst that can
utilize visible light would be more effective because UV light
accounts for only 7% of sunlight, while visible light accounts for
47%, and the remaining energy falls in the nearꢀinfrared (IR) region
of the electromagnetic spectrum. Some semiconductors with reduced
band gaps obtained by modifying their valence state of the metal
dyeꢀsensitized hydrogen production in water with electron donor
system, the organic dye absorbs the visibleꢀlight component of
sunlight and becomes excited, leading to injection of an electron into
the conduction band of the semiconductor, which then enables
hydrogen production from waterꢀcontaining electron donor (with a
coꢀcatalyst for increasing the hydrogen production yield). Dyes with
suitable absorption and HOMOꢀLUMO energies can be selected by
modifying their electronic structures to match the conduction
band/valence band level of the photocatalyst. A variety of dyes for
hydrogen production from water decomposition have been
developed, including a Pt/TiO2/dye system with a ruthenium
complex, 8 porphyrin, 9 coumarin, 10 cyanine, 11 and xanthene12 as the
dye. To improve the hydrogen production yield in dyeꢀsensitized,
waterꢀsplitting cell systems, an understanding of the mechanism of
the water decomposition reaction is needed. To improve the
hydrogen production efficiency,
a reduction in holeꢀelectron
recombination between the dye and TiO2, and electron trapping after
injection of the electrons from dye at the excited state, using
e.g.,water or a sacrificial reagent, is an important issue. Kim et al.
reported that a tinꢀporphyrin dye showed radical species lasting long
enough (< 30 ꢀs) on TiO2 for hydrogen production.24 We have
cations or through doping have exhibited visibleꢀlightꢀdriven
photocatalytic waterꢀsplitting activity.
3,4
For example, the
semiconductor Ta3N5 exhibited waterꢀsplitting activity when
exposed to light of wavelengths up to 600 nm. 5
This journal is © The Royal Society of Chemistry 2013
J. Name., 2013, 00, 1-3 | 1

Journal of Materials Chemistry A
Page 2 of 9
View Article Online
ARTICLE
Journal Name
DOI: 10.1039/C5TA04991A
studied the spacer effect of benzo[b] phenothiazineꢀcontaining THF solutions. The lowest energy bands of the dyes were observed
14
dyads13 on hydrogen production.
The spacer is considered to at
decrease the charge recombination rate between the electron donor 457–460 nm for C6, C16 and C22, while that of the dye C1 was
and the semiconductor, leading to enhancement of the hydrogen slightly blueꢀshifted (Fig 2 left). These lowest energy bands were
production amount. Han et al. reported that a hydrophilic substituent assigned to the characteristic chargeꢀtransfer (CT) bands between the
on the dye led to a decrease of the charge recombination rate donor carbazole and acceptor cyanoacrylic acid groups in the dyes.
because of interactions between the hydrophilic group and the water The fluorescence spectra of the dyes in THF solutions included
15
medium that in turn led to relaxation at the excited state. This peaks at 549–553 nm for C6, C16 and C22, while the peak for C1
longꢀlived charge recombination improves the hydrogen production. was once again slightly blueꢀshifted (542 nm). Based on the 0ꢀ0
Lee et al. also reported that the introduction of an alkyl chain in offset bands, it was determined that all of the dyes had similar
phenothiazine improved the hydrogen production performance. energy band gaps (2.39–2.40 eV) and HOMOꢀLUMO energies.
Although the reason for the effect of the alkyl chain on the yield of
hydrogen production was unclear, they proposed that the orientations
of the dyes on the TiO2 were changed and resulted in acceleration of
smooth electron injection, reduction of the decomposition reaction
rate during the excited state of the dye and electron injection into the
semiconductor and/or regeneration of the dye using sacrificial
reagents. 16 Other factors affecting the yield of hydrogen production
include the temperature, solvent medium and the stability of the
photocatalyst. In the present study, we investigated the visibleꢀlightꢀ
driven water decomposition hydrogen production of carbazoleꢀ
based, metalꢀfree organic dyes with alkoxy substituents on the
Scheme 1. Synthetic scheme of dyads C1-C22
.
carbazole moiety with varying chain lengths. The carbazole moiety
is one of the good electron donor moieties in the donorꢀspacerꢀ
acceptor systems such as in dyeꢀsensitized solar cells.40 In particular,
the hydrophobic effect of the alkxoy chains on the hydrogen
production yield was evaluated (Fig.1). A 207% increase in the
hydrogen production yield was observed for the dye with the longest
alkoxy chain compared to that for the dye with the shortest alkoxy
substituent.
i) K2CO3, DMSO, Lꢀproline, CuI, 1ꢀbromoꢀ4ꢀalkoxybenzene. 9ꢀ51%;
ii) POCl3, DMF. 41ꢀ57%; iii) NaBH4, THFꢀMeOH, then iv)
triphenylphosphane hydrobromide, toluene. 67ꢀ98% in 2 steps; v)
[2,2':5',2''ꢀterthiophene]ꢀ5,5''ꢀdicarbaldehyde, K2CO3, 18ꢀcrownꢀ6ꢀether,
DMF. 64ꢀ78% vi) cyanoacetic acid, ammonium acetate, AcOH.; vii)
cyanoacetic acid, ammonium acetate, THFꢀAcOH. 34ꢀ67%.
Cyclic voltammetry
To examine the HOMO and LUMO energies of dyes C1–C22, cyclic
voltammetry experiments were performed (Fig. 2 right). One
ꢃꢄ ꢅ ꢇ
reversible electron oxidation at ꢀꢁ/ꢂ
observed for C1, while the ꢀꢁ/ꢂ ꢆ ꢈ values for C6, C16 and C22
ꢆ = 1.17 V (vs. NHE.) was
ꢃꢄ ꢅ ꢇ
were all found to be 1.16 V (vs. NHE). However, no irreversible
reduction potentials were observed to −1.0 V in the THF potential
window (V vs NHE). The LUMO energies of dyes C1–C22 were
estimated to be −1.23 to −1.24 V (V vs NHE) from the 0ꢀ0 offset
bands that were estimated from overlap position between absorption
and fluorescence spectra. In addition, only a very small difference in
the oxidation potentials was found for C1 and the other three dyes
(C6, C16 and C22), indicating that the electron donating ability of
the alkoxy chain to the dye backbone can be neglected, particularly
for the dyes with long alkoxy chains, i.e. C6, C16 and C22.
Fig. 1 Series of synthesized dyes C1-C22
.
Results and Discussion
Synthesis
Dyes C1, C6, C16 and C22 were synthesized from 9Hꢀcarbazole (1)
(Scheme 1). The Nꢀposition of the carbazole was substituted with 4ꢀ
alkoxyphenyl groups via coupling with 1ꢀbromoꢀ4ꢀ(alkoxy)benzene
derivatives in the presence of K2CO3 using CuI/Lꢀproline as the
catalyst. The desired alkoxybenzene carbazoles (2a–d) were
obtained in 9%–51% yield. The carbazoles (2a–d) were then
subjected to VilsmeierꢀHaack reaction conditions to afford the
corresponding aldehydes (3a–d) in 41%–57%. Subsequently, the
aldehydes (3a–d) were reduced using NaBH4 and then reacted with
triphenylphosphine hydrobromide to give the Wittig salts (4a–d) in
67%–98% over the two steps. The Wittig salts (4a–d) were then reacted
with terthiopheneꢀdicarboxyaldehyde to afford aldehydes (5a–d) in
64%–78%. Finally, these aldehydes were converted via Knoevenagel
condensation to donor (carbazole)ꢀspacer (terthiophene)ꢀacceptor
(cyanoacrylic acid) moieties (C1–C22) in 34%–67%. Further
Figure2. Absorption spectra of dyes C1–C22 in THF (3.0 × 10ꢀ4 M)
(left), Cyclic voltammograms of C1–C22 in 0.1 M Bu4NPF6ꢀ
containing THF solutions. The scan rate was 100 mV/s. (right)
†
synthetic details are explained in the ESI .
Physical properties
Absorption and fluorescence spectra
The absorption spectra of the dyes were obtained for 3.0 × 10−5
M
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 2012

Page 3 of 9
Journal Name
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C5TAA04R9T9I1CALE
Table 1. Experimental and calculated parameters of all dyes.
max nma
Comp.
FL (nm)
0ꢀ0 band HOMO/LUMOb Comp abs (f)d
Comp HOMO/LUMO
λ
(ε/Mꢀ1cmꢀ1)
438 (42700)
459 (46700)
(nm) / eV
(V vs NHE)c
(eV)d
542
549
517 / 2.40
516 / 2.40
1.17/ꢀ1.24
1.16/ꢀ1.24
499.88 (2.0766)
501.64 (2.0864)
ꢀ5.79/ꢀ2.08
ꢀ5.77/ꢀ2.08
C1
C6
460 (46700)
457 (47000)
553
550
518 / 2.39
517 / 2.40
1.16/ꢀ1.23
1.16//ꢀ1.24
501.80 (2.0967)
501.70 (2.0937)
ꢀ5.77/ꢀ2.08
ꢀ5.77/ꢀ2.08
C16
C22
a : Measured in THF, b : LUMO calculated by HOMO + E0ꢀ0
d : Obtained by TDDFT at the BhandLYP/6ꢀ31G(d) level
c : Calibrated according to Fc/Fc+ = ꢀ 0.63 V vs. NHE.
,
Theoretical computations for molecules
The absorption, fluorescence and CV spectra of all of the dyes
indicated that they had similar photophysical and electrochemical
properties. The optimized structures of C1-C22 were obtained by
B3LYP/6ꢀ31G(d) basis set. Timeꢀdependent density functional
theory (TDꢀDFT) calculations were performed for the HOMO and
LUMO energies at the B3LYP17 or BHandHLYP17 or M0626
functions with the 6ꢀ31G(d) basis set. All of the trends of HOMOꢀ
LUMO energies and gaps, and oscillator strength were the same as
the experimental data trends. In these theoretical results,
BHandHLYP showed much closer energy gaps with experimental
The trends in the calculated lowest energy bands, oscillator strengths
and HOMOꢀLUMO energies agreed well with those observed
experimentally. These results indicated that the alkoxy groups had
little electronꢀdonating or electronꢀaccepting abilities, but the
absorption structure and HOMOꢀLUMO gaps were not changed
(Fig. 3). The physical and theoretical properties are summarized in
Table 1.
Hydrogen production
†
Visibleꢀlightꢀdriven (>420 nm, 300
W Xe lamp) hydrogen
(Table S1 ), and results suggested that nearly all of the HOMOs and
production was evaluated using a dye (C1–C22)/ TiO2/ 1 wt% coꢀ
catalyst system in water (10% triethanolamine (TEA), pH = 7). All
of the reactions were performed for 24 h. The hydrogen production
performance with several coꢀcatalysts, i.e. Pt, Au, RuO2, NiO, and
without a coꢀcatalyst (w/o) was compared using dye C16 (Figure
LUMOs were delocalized on the donorꢀterthiophene and
terthiopheneꢀcyanoacrylic acid moieties, respectively. However, no
contribution to the HOMOꢀLUMO levels was found for the
alkoxybenzene moieties. The lowest energy absorption bands for all
of the dyes were attributed to the HOMOꢀLUMO levels, suggesting
chargeꢀtransferꢀtype interactions. For the dye C1, the lowest energy
band and HOMOꢀLUMO levels were estimated to be approximately
500 nm and −5.79/ −2.08 eV, respectively, while the values for the
other three dyes were approximately 502 nm and −5.77/ −2.08 eV,
respectively.
†
†
S1 and Table S2 ). We found that the order of hydrogen production
was Pt (1405 ꢀmol) > Au (142 ꢀmol) > RuO2 (66 ꢀmol) > NiO (56
ꢀmol) > w/o (9 ꢀmol). The results suggested that the extraction of
electrons from the dye could be promoted by the coꢀcatalyst. The
trends can be explained by the order of the work functions: Pt (5.65ꢀ
5.32 eV)27 > Au (5.10 eV) 27a > RuO2 (4.8 – 5.04 eV) 28 > NiO (3.8 –
5.0 eV) 29, and no help of electron promotion (w/o), because a larger
work function can promote the electron acceptor properties.30
1600
C1
1400
C6
1200
1000
800
600
400
200
0
C16
C22
0
4
8
12
16
20
24
Time (h)
Figure4. Photocatalytic activities of dyes C1-C22 in hydrogen
production from water. Reaction conditions: 10 vol% aqueous
TEA (10 mL), 33.0 mg dye/ TiO2 /Pt catalyst, pH = 7.0.
Figure3. Potential energy levels of dyes and generated
molecular orbitals of HOMO and LUMO geometries from the
optimized structure of C1–C22 at the BHandHLYP/6ꢀ31G(d)
level.
This journal is © The Royal Society of Chemistry 2012
J. Name., 2012, 00, 1-3 | 3

Journal of Materials Chemistry A
Page 4 of 9
View Article Online
ARTICLE
Journal Name
DOI: 10.1039/C5TA04991A
Table 2. Photocatalytic performancea, decay lifetimes, and
impedance results (resistances) of all dyes on TiO2
Comp.
Dyeꢀloading
ꢀ
H2 amount
R1
(ohms)
R2
(ohms)
τ
τ
τ
3
1
2
(
mol /33 mg) (ꢀ
mol)b / TONc
(ps)
1.07
1.79
1.42
1.06
(ps)
8.52
11.8
16.7
10.1
(ps)
89.2
99.0
127.0
111.2
0.927
0.904
0.959
0.872
694 / 1497
11.9
11.1
11.5
10.9
205.2
222.7
293.4
249.3
C1
C6
C16
C22
1048 / 2319
1405 / 2930
1349 / 3094
a: A 300 W Xe lamp was used and wavelengths below 420 nm were
cut off using an optical fillter. b: H2 produced after 24 h.
c : TON = (2
× amount of H2 / dye loading amount after 24
h)/amount of dye.
TiO2 systems.14 Han et al. have reported that the electron injection
from dyes into TiO2 sites occurs on a time scale of hundreds of
femtoseconds. A long charge recombination time of greater than 5 ps
would allow electron injection from the dye into the conduction band
of TiO2, while species remaining less than 5 ps of would indicate a
charge recombination process between dye and TiO2 at the surface
and/or shallow traps on TiO2.15, 19
In dyeꢀsensitized solar cell systems, it has been reported that when
the angle of orientation of the dye on the TiO2 surface is large, the
charge recombination lifetime and efficiency decrease dramatically
because of throughꢀspace charge recombination between the dye and
the TiO2 surface.20, 21
The effects of dyes were further examined with a dye (C1–C22)/
TiO2/ 1 wt% Pt system in water (10% triethanolamine (TEA), pH =
7). The highest hydrogen production yield was observed for C16,
which exhibited an amount of the hydrogen production of 1405
ꢀmol, and the hydrogen production yield was found to differ in the
order C16 (1405 ꢀmol) > C22 (1349 ꢀmol) > C6 (1048 ꢀmol) > C1
(694 ꢀmol). After correcting for the number of dye molecules
loaded on the photocatalyst, the turnover numbers were found to be
1497 for C1, 2319 for C6, 2390 for C16 and 3094 for C22 (Fig.4).
In dye sensitized photocatalytic hydrogen production, the extinction
coefficient, absorption range of dye, and HOMOꢀLUMO energy
levels of the dye are important properties to modify in order to
improve the absorption of the light energy. The LUMO energy
positions affect the driving force of electron injection from the exited
dye to the CB of TiO2. The absorption properties of the various dyes
were quite slimmer. Furthermore, the LUMO energies of the dyes
were similar, relative to the CB of TiO2 (Fig 3). These results
suggested that the effect of the electronic properties of the dyes does
not account for the difference yields of hydrogen production. Lee et
al. reported that alkyl chain substituents on phenothiazine dyes used
in dyeꢀsensitized water decomposition cells may improve the
hydrogen production yield because of suppression of the rate of
decomposition of the dye itself and/or its reaction with water or a
change in the orientation of the dye on the TiO2 surface.16 As can be
seen in Fig.4., all the hydrogen production decay profiles of the four
dyes were quite similar, which suggested that the alkoxy chains did
not influence the decomposition pathway. It is considered that
hydrophilic substituents on the dyes would improve the charge
Figure 5. Solid state absorption spectra of dyeꢀloaded (C1-C22) TiO2
recombination lifetime of the excited states, while hydrophobic film.
groups would not be expected to have a strong influence in water.
The lifetimes of the four dyes investigated in the present study were
Time-resolved absorption spectra
found to fall in the order of C16 (16.7 ps) > C6 (11.8 ps) ≧ C6 (10.1
ps) > C1 (8.52 ps) for τ2, and the radical cation species survived in
the order of C16 (127.0 ps) > C22 (111.2 ps) > C6 (99.0 ps) > C1
(89.2 ps) for τ3. Han et al. reported that a hydrophilic substituent on
the dye led to a dramatic decrease of the charge recombination rate
by relaxation with water..15 In our results, no change of the structure
of signal positions in the timeꢀresolved absorption spectra at the
range of 0 to 150 ps was confirmed. This result suggested that the
effect of the alkoxy chain length for relaxation with water solvent
was small.
To elucidate the reason for the improvement in the TON with
increasing chain length of the alkoxy substituent, timeꢀresolved
absorption spectra were recorded. The spectra were obtained using a
dyeꢀloaded TiO2 film (approximately 8–10 nm thick on a quartz
plate) that was dipped in pure water (pH = 7). It can be seen in Fig.5
that the solidꢀstate absorption spectra for the dyes on TiO2 were
similar.
The timeꢀresolved absorption spectra of C22 on the TiO2 film in
water (pH=7) revealed that a rise of a broad peak (600 to 750 nm)
maximized around 690 nm was observed on a time scale of 1 ps or
even longer. It exhibited absorption above approximately 600 nm for
the radical cation species, while a negative peak was observed below
600 nm induced by groundꢀstate bleaching (Fig. 6a). 18 In addition,
all the dyes were found to exhibit triple exponential decay functions,
and the time decay constants for compounds C1–C22 fell in similar
ranges (Fig.6b), i.e. τ1 = 1.07–1.79 ps, τ2 = 8.52–16.7 ps and τ3 =
89.2–127.0 ps. These decay lifetimes are in good agreement with
those observed previously for oligoꢀthiopheneꢀcontaining dyeꢀloaded
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 2012

Page 5 of 9
Journal Name
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C5TAA04R9T9I1CALE
(MD) simulations were performed for the organic dye attached
to the anatase surface and 50 water molecules. The early stage
structure of C1 used the structure optimization gave. The early
stage structure of C6 used the one which changed the alkoxy
group of the optimization structure of C1 to a hexyloxy group.
The simulations were performed with PBE theory implemented
in the VASP program and 300 eV cutꢀoff energy. The time step
was 1.0 fs and the simulation time was 1.8 ps for C1 on TiO2.
The dye C6 extended calculation until 2.5 ps to check influence
of a hexyloxy group. The simulation temperature was kept
constant at 300 K. The snapshots are shown in Figure 7.
Although during the MD simulation, the carbazole moiety on
the molecule and its alkoxy chain were periodically rotating,
the dye molecules were perpendicular to the surface and there
were only insignificant deviations from the early stage
geometry, suggested that the dye was keeping the standing
Figure 6. (a) Typical timeꢀresolved absorption spectra of C22-loaded
TiO2 film in water (pH = 7) condition, and (b) time decay profile of
dyeꢀloaded (C1-C22) TiO2 at 690 nm in the range of ꢀ10 to 150 ps.
The orientation of the dye on TiO2 is controlled by the connecting position at 300 K with water molecules.
mode of dye to TiO2, and the perpendicular structure is favourable to
reduce the charge recombination lifetime.34, 35
Impedance and Incident Photon to Current Conversion
Efficiency (IPCE) spectra
Dye-TiO2 Surface Calculation Results
To gain further insight into the influence of the alkoxy chain on
improving the charge recombination and hydrogen production
amount, electrochemical impedance spectra (EIS) of the dyeꢀloaded
TiO2 films were obtained using a three electrode cell (working
electrode/reference electrode/counter electrode) system.
Calculations of the dye on the TiO2 surface were performed using
density functional theory with plane wave basis set implemented in
the Vienna Abꢀinitio Software Program (VASP) with PBE
functional.31ꢀ33
The dyeꢀloaded TiO2 film coated on fluorineꢀdoped tin oxide
(FTO) glass was used as the working electrode, a Pt mesh was
used as the counter electrode, and Ag/AgCl was used as the
reference electrode. An aqueous Co2+/Co3+ solution was used as
the electrolyte. The small, high frequency semicircles in the
range of 1–20 ohms seen in the inset of Fig.8 (left) have been
attributed to the electrode/electrolyte impedance. On the other
hand, major semicircular arcs related to charge recombination
at the interface between the TiO2 and dye/electrolyte were
observed in the low frequency range (~20–300 ohms
resistance).22 The diameter of the largest semicircles decreased
in the order C16 (293.4 ohms) > C22 (249.3 ohms) > C6 (222.7
ohms) > C1 (205.2 ohms). These results are in reasonable
agreement with the results obtained for the yields of hydrogen
production; it is possible that the water medium trapped the
electrons when the hydrophobicity was decreased. Although the
length of the alkoxy chain in C22 was greater than that for the
Figure 7. First principle molecular dynamics results of (a, b) C1 and (c,
d) C6 on TiO2 (101) surface with 50 water molecule. The snapshots at other dyes, its hydrogen production and charge recombination
(a) 0 ps for C1 on TiO2, (b) after 1.8 ps, and (c) 0 ps for C6 onTiO2, (d)
after 2.5 ps.
rates were lower and faster, respectively, than those for C16
However, the amount of dye in C22 loaded onto the TiO2
(0.872 mol/33 mg TiO2) was the lowest of all of the four dyes.
These results, therefore, suggested that the packing structure for
C22 on the TiO2 surface was not well defined because of the
longer length of the alkoxy chain. They may also provide an
indication of the orientation of C22 on the TiO2. If the electron
trapping was suppressed at the interface of solvent and
.
ꢀ
During the geometry optimization a 300 eV cutꢀoff energy was
employed with Monkhorstꢀ Pack automatic 1
sampling. The periodic cell has dimensions of 22.7 Å
50 Å and contains 479 atoms for C1 with TiO2. The TiO2
was modelled with three layers of the anatase surface with
×
2
×
1 kꢀpoints
×
10.2 Å
×
(101) Miller indices. All surface oxygen atoms were hydrogen dye/TiO2, because of the alkoxy chain, the overall efficiency of
terminated in order to avoid a surface induced dipole moment.
During the optimization the coordinates of the middle layer
were kept fixed while the coordinates of the top and bottom
electron correction from the dye to the TiO2 surface would be
increased.35, 36
To determine the overall light to current efficiency, IPCE
layers were fully relaxed. Due to the high cost of computation spectra were obtained for the dyeꢀloaded TiO2 films and are
when a longꢀalkyl chain is attached, we examined the molecule
C1 for optimization. The organic dye was connected at the
surface to Ti atoms through TiꢀO bonds at bidentate geometry
shown in Fig 8 (right) over the range from 350–750 nm. IPCE
results combined the lightꢀtoꢀcurrent efficiency from light
harvesting, electron injection efficiency, and electron correction
with 82.9o for C1 on TiO2 of dihedral angle between TiO2 efficiency.37 Although the IPCE cannot evaluate these
surface and cyanoacrylic acid moiety, suggested perpendicular
components separately, over the entire range, the charge
transfer efficiencies followed the order C16 C22 C6 C1
to the surface. (Figure S2†)
>
>
>
,
To understand the dynamics of the organic dyes on the TiO2
surface in water medium, first principles molecular dynamics
confirming that dye C16 exhibited the highest photonsꢀtoꢀ
current efficiency. These results also indicated that the length of
This journal is © The Royal Society of Chemistry 2012
J. Name., 2012, 00, 1-3 | 5

Journal of Materials Chemistry A
Page 6 of 9
View Article Online
ARTICLE
Journal Name
DOI: 10.1039/C5TA04991A
1
the alkoxy chain is important and should be designed to achieve The H and 13C NMR were recorded on a BrukerAV600 (600
a good hydrophobic effect for dyeꢀsensitized hydrogen MHz). The 1H and 13C NMR chemical shifts were reported as d
production reactions from water.
values (ppm) relative to external Me4Si. The coupling constants
(J) were given in hertz. MALDI mass spectra were recorded on
a
Bruker Auto flex. The sample was dissolved in
dichloromethane first, then equal amounts of this sample
solution (1.0 m L) and a dithranol in dichloromethane (DIT, 1.0
m L) were spotted onto the stainless steel sample plate for
injection together. High resolution mass spectra were recorded
on a JEOL LMSꢀHXꢀ110 spectrometer. FAB MS spectra were
recorded with 3ꢀnitrobenzyl alcohol (NBA) as the matrix.
Analytical thin layer chromatography (TLC) was performed on
Silica gel 60 F 254 Merck. Column chromatography was perꢀ
formed on KANTO Si60N (neutral). Absorption spectra were
recorded on a SIMAZU IR spectrophotometer. IR spectra were
recorded on a SHIMAZU spectrometer. Redox potentials were
carried out on a BASꢀ100B/W electrochemical analyser. CV
measurements were performed using a cell equipped with
glassy carbon as the working electrode, platinum wire as the
counter electrode, and Ag/AgNO3 as the reference electrode.
All electrochemical measurements were performed under an Ar
atmosphere at room temperature in THF solution (5 x 10 ꢀ4 M)
Figure 8. EIS Nyquist plot of dyeꢀloaded TiO2 at 0 V bias under
illumination (left). IPCE plots of dyeꢀloaded TiO2 (right).
Conclusions
Novel donor–spacer–acceptor dyads with 4ꢀalkoxybenzene
substituents on the donor carbazole units were synthesized for
investigation of the hydrophobic effect of the alkoxy chains in
dyeꢀsensitized hydrogen production reaction from water. Their
properties were in good agreement with the values obtained
using DFT calculations. While barely any changes in the
HOMOꢀLUMO gaps and electronꢀdonating abilities of dyads
C1–C22 were observed, the visibleꢀlightꢀdriven photocatalytic
hydrogen production of dye/ TiO2/Pt catalysts at pH 7 in 10%
containing
0.10
M
tetraꢀnꢀbutylammoniumhexafluoroꢀ
phosphate (Bu4NPF6) as a supporting electrolyte at a scan rate
of 100 mV s ꢀ 1. The ferrocene/ferrocenium (Fc/Fc+) couple was
used as an internal standard. THF was distilled from sodium
benzophenone ketyl. Toluene was distilled from CaH2. Other
solvents and reagents were of reagent quality, purchased
commercially, and used without further purification.
aqueous TEA differed for the four dyes. The hydrogen Dye-Loading process into 1wt% Pt loaded TiO2 anatase
production yield increased as the length of the alkoxy chain The procedure of 1.0 wt% Ptꢀloaded TiO2 was as follows in
accordance with the literature.23 The commercially available
increased, and the highest hydrogen production yield (measured
TiO2 source (1.0 g, <25 nm, anatase) was added to MeOH (25
mL) and H2PtCl6 (0.25 mL, 8 wt%). The mixture was irradiated
as the TON) was observed for C22 (3094), while the lowest
yield was found for C1 (1497). Overall, a 207% increase in the
hydrogen production yield was observed because of the using a Xe lamp (300 W) for 25 min, leading to the production
of TiO2 – Pt nanoparticles. The TiO2 – Pt nanoparticles were
separated by centrifugation (3600 rpm, 20 min); the separated
incorporation of a hydrophobic substituent into the dyads. On
the other hand, the hydrophobic substituents did not improve
the stability of the catalysts, and the deactivation times for all
the compounds were approximately 24 h. In addition, the timeꢀ
resolved absorption spectra for all the dyes on TiO2 exhibited
tripleꢀdecay components, and the longest decay component
behaved in a similar manner to that of the hydrogen production
yield for the four dyes, suggesting that the charge
recombination lifetime was affected by the different alkoxy
chain lengths. EIS analyses indicated that the dye C16 had the
longest charge recombination lifetime between the TiO2 and
dye/electrolyte, while that of C1 was the shortest. Finally, the
IPCE spectra indicated the same trend as observed in the EIS
plots, suggesting that the alkoxy chain enhanced electron
injection from the dye into the TiO2. Furthermore, the hydrogen
production yield for dye C22 with the longest alkoxy group was
nearly saturated, as was that for C16, because the dyeꢀpacking
structures on the TiO2 in these two systems was not well
defined because of the longer alkoxy chains. This reduced
packing order on the TiO2 enabled easy contact with the water
at the water/TiO2 interface. Therefore, we have demonstrated
that hydrogen production via visibleꢀlightꢀdriven dyeꢀsensitized
TiO2
– Pt nanoparticles were filtered, washed with MeOH
successively, and then dried under vacuum, resulting in the
production of Ptꢀloaded TiO2 nanoparticles.
1.0 wt% Ptꢀloaded TiO2 (100 mg) was immersed in a MeOH
solution containing 3
×
10ꢀ 4M dye sensitizers (10 mL) for 24 h
in the dark. The solvent was removed by centrifuged (3600 rpm,
20 min) and filtration, rinsed with anhydrous MeOH. The
powder of dyeꢀloaded TiO2 catalyst was dried under vacuum. It
was used for hydrogen production reaction without further
treatment.
Hydrogen Production Reaction
The photocatalytic hydrogen production experiment was
performed with a conventional closed circulating system with a
dead volume of approximately 500 mL. The dye/TiO2 /Pt
catalyst (33 mg) was suspended in 10 mL of 10 v/v%
triethanolamine water (pH = 7.0, adjust with HCl aq.). A quartz
reaction cell was irradiated using an external light source
comprising a 300 W Xe lamp (Ushio Inc., Japan) with <420 nm
hydrogen production can be increased through the use of cut o
ff glass filter. During the H2O photochemical reaction, the
hydrophobic effects at the interface of the photocatalyst and the reaction mixture was mixed using a magnetic stirring bar. The
water medium.
amount of H2 gas was measured with a TCD gas
chromatograph (GCꢀ8A, Shimadzu Corp., Japan), which was
connected to a conventional volumetric circulating line with a
vacuum pump.
Experimental Section
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 2012

Page 7 of 9
Journal Name
Journal of Materials Chemistry A
View Article Online
ARTICLE
DOI: 10.1039/C5TA04991A
Culture, Sports, Science and Technology (MEXT), Japan, and
was performed under the Cooperative Research Program of
"Network Joint Research Centre for Materials and Devices
(IMCE, Kyushu University)". MW, AS, SRB, and NHP
acknowledge support from I2CNER, funded by the World
Premier International Research Centre Initiative (WPI),
Ministry of Education, Culture, Sports, Science and
Technology of Japan (MEXT), Japan.
Time transient absorption measurements
The titaniaꢀoxide pastes of TiꢀNanoxide T/SP that was purchased
from Solaronix. The quartz plate (transmission > 90% in the visible)
was used as substrate. A thin film of TiO2 (8–10 mm) was coated on
a quartz substrate. It was immersed in an anhydrous THF solution
containing 3 x 10ꢀ4 M dye sensitizers for at least 24 h, then rinsed
with anhydrous THF and dried. Time transient absorption spectra of
dyes on thinꢀfilm or in solution were performed in an ambient
atmosphere. The femtosecond transient absorption data were
collected with a pump and probe transient absorption spectroscopy
system (Ultrafast Systems, Helios). The pump light (400 nm) was
used from an amplified Tiꢀsapphire laser system (SpectraꢀPhysics,
Solsticeꢀ100F, < 100 fsec, 5 kHz). The UV/Vis range (350ꢀ750 nm)
was detected with a linear CCD array (Ocean Optics, S2000). The
triple exponential decay model was used for fitting. (eq. 1)
Notes and references
a International Institute for CarbonꢁNeutral Energy Research
(I2CNER),Kyushu University, 744 Motooka, Nishiꢁku, Fukuoka 819ꢁ0395,
Japan
b Department of Applied Chemistry, Kyushu University, 744 Motooka, Nishiꢁ
ku, Fukuoka, 819ꢁ0395, Japan
c Department of Chemistry, Tung Hai University, Taichung 407, Taiwan
∆ꢉꢊ/∆ꢉꢊꢋ ꢈ ꢌꢁꢍꢎꢏ ꢐ− ꢑ ꢔ + ꢌꢂꢍꢎꢏ ꢐ− ꢑ ꢔ + ꢌꢖꢍꢎꢏ ꢐ− ꢑ ꢔ (1)
|
|
†
Electronic Supplementary Information (ESI) available: [Synthetic
ꢒꢓ
ꢒꢕ
ꢒꢗ
detail of dyes (C1ꢀC22), and 1H and 13C NMR spectra, hydrogen
Fabrication of Dye-loaded TiO2 coated FTO glass
production test with different coꢀcatalyst.]. See DOI: 10.1039/b000000x/
Dyeꢀloaded TiO2 coated FTO glass was used as working
1. A. J. Bard and M. A. Fox, Acc. Chem. Res., 1995, 28, 141.
2. C. Song, Catal. Today, 2006, 115, 2.
3. M. Ni, M. K. H. Leung, D. Y. C. Leung and K. Sumathy, Renewable
Sustainable Energy Rev., 2007, 11, 401.
4. A. Kudo and Y. Miseki, Chem. Soc. Rev., 2009, 38, 253.
5. M. Hihashi, K. Domen, and R. Abe, Energy Environ. Sci., 2011, 4,
4138.
6. W. J. Youngblood, S.ꢀY. A. Lee, K. Maeda and T. M. Mallouk, Acc.
Chem. Res., 2009, 42, 1966
7. J. R. Swierk and T. E. Mallouk, Chem. Soc. Rev., 2013, 42, 2357.
8. (a) E. Borgarello, J. Kiwi, E. Perlizzetti, M. Visca and M. Grätzel,
electrode, which has 3 x 2 cm dimension of FTO glass with
8
ꢀ/sq as surface resistivity, and has 7 x 7 mm dimension (8 um
thick) of TiO2 layer. The TiO2 layer (TiꢀNanoxide T/SP,
SOLARONIX) was screenꢀprinted on FTO glass. The TiO2ꢀ
o
coated FTO glass was preꢀdried 1 h at 100 C, followed by
heated 80 min at 500 oC in air condition. It was immersed in an
anhydrous THF solution containing 3 x 10ꢀ4 M dye sensitizers
for at least 24 h, then rinsed with anhydrous THF and dried, to
give Dyeꢀloaded TiO2 coated FTO glass.
Nature, 1981
and T. Sakata, J. Phys. Chem., 1982, 88, 4516; (c) Y. I. Kim, S. J.
Atherton, E. S. Brigham and T. E. Mallouk, J. Phys. Chem., 1993
97, 11802; (d) K. Hirano, E. Suzuki, A. Ishikawa, T. Moroi, H.
Shiroishi and M. Kaneko, J. Photochem. Photobiol., A, 2000, 136
, 289, 158; (b) T. Kajiwara, K. Hashimoto, T. Kawai
Electrochemical Impedance Spectra (EIS) measurements
The experimental configuration for the EIS measurements
consisted of a working electrode/ reference electrode / counter
electrode system. Dyeꢀloaded TiO2 coated FTO glass was used
as working electrode. Pt mesh was used as counter electrode.
The reference electrode used was Ag/AgCl. An aqueous
,
,
157; (e) K. Maeda, M. Eguchi, W. J. Youngblood and T. E.
Mallouk, Chem. Mater., 2008, 20, 6770; (f) X. Zhang, U. Veikko, J.
Mao, P. Cai and T. Peng, Chem.–Eur. J., 2012, 18, 12103; (g) H.
Luo, W. Song, P. G. Hoertz, K. Hanson, R. Ghosh, S. Rangan, M.
K. Brennaman, J. J. Concepcion, R. A. Binsteas, R. A. Bartynski, R.
Lopez and T. J. Mayer, Chem. Mater., 2013, 25, 122.
9. W. Kim, T. Tachikawa, T. Majima, C. Li, HꢀJ. Kim, and W. Choi,
Energy Environ. Sci., 2010, 3, 1789.
10. R. Abe, K. Sayama and H. Arakawa, J. Photochem. Photobiol., A,
2004, 166, 115.
11. R. Abe, K. Shinmei, K. Hara and B. Ohtani, Chem. Commun., 2009,
3577.
solution of 0.16
M
Co[(bpy)3](NO3)2
,
0.032
M
Co[(bpy)3](NO3)2, and 0.08 M KCl served as the electrolyte.
EIS measurements were performed using a Novocontrol Alpha
o
analyzer with a ZG4 interface at 0V bias voltage at 59
±
2 C,
with a frequency range from under roomꢀlight illumination. The
equivalent circuit is Gerischer element22 and analysis was
performed by Zꢀview software.
12. (a) T. Shimizu, T. Iyoda and Y. Koide, J. Am. Chem. Soc., 1985, 107,
35; (b) X. Zhang, Z. Jin, Y. Ki, S. Li and G. Lu, J. Phys. Chem. C,
2009, 113, 2630; (c) P. Chowdhury, H. Gomaa and A. K. Ray, Int.
J. Hydrogen Energy, 2011, 36, 13442.
13. M. Watanabe and T. Ishihara, Acta Cryst., 2014, E70, o1026.
14. M. Watanabe, H. Hagiwara, A. Iribe, Y. Ogata, K. Shiomi, A.
Staykov, S. Ida, K. Tanaka and T. Ishihara, J. Mater. Chem. A,
2014, 2, 12952.
IPCE measurement
For the measurement of IPCE was carried out by the structure
of Dyeꢀloaded TiO2 coated FTO/electrolyte / Pt system. The
electrolyte was used by LiI (0.05 M), I2 (0.05 M), and TBP (0.5
M) in MeCN at 25oC.
Theoretical Computation
15. W.ꢀS. Han, K.ꢀR. Wee, H.ꢀY. Kim, C. Pac, Y. Nabetani, D.
Yamamoto, T. Shimada, H. Inoue, H. Choi, K. Cho and S. O. Kang,
Chem.–Eur. J., 2012, 18, 15368.
16. J. Lee, J. Kwak, K. C. Ko, J. H. Park, J. H. Ko, N. Park, E. Kim, D.
H. Ryu, T. K. Ahn, Y. J. Lee and S. U. Son, Chem. Commun., 2012,
48, 11431.
17. A. D. Becke, J. Chem. Phys., 1993, 98, 1372.
18. R. Bereram, R. V. Grondelle and J. T. M. Kennis, Photosynth. Res.,
2009, 101, 105.
Geometry optimization of the dyes were accomplished by
Gaussian G09 (Rev. C. 01) program38, and dye chemisorbed on
the anatase surface was performed using the DFT implemented
in the VASP program with PBE functional. Throughout this
study, we have used the graphical visualization package
VESTA to analyze and visualize the computed DFT results.39
19. A. furube, T. Asahi, H. Masuhara, H. Yamashita and M. Anpo, J.
Phys. Chem. B 1999, 103, 3120.
20. A. S. Hart, C. B. KC., H. B. Gobeze, L. R. Sequeria, and F. D’Souza,
ACS Appl. Mater. Interfaces, 2013, 5, 5314.
Acknowledgements
This work was supported by a GrantꢀinꢀAid for Science
Research (YB 26810107) from the Ministry of Education,
This journal is © The Royal Society of Chemistry 2012
J. Name., 2012, 00, 1-3 | 7

Journal of Materials Chemistry A
Page 8 of 9
ARTICLE
VJieowuArrtnicalelONnlainme e
DOI: 10.1039/C5TA04991A
21. (a) H. Imahori, S. Kang, H. Hayashi, M. Haruta, H. Kurata, S. Isoda,
S. E. Canton, Y. Infahsaeng, A. Kathiravan, T. Pascher, P. Chábera,
A. P. Yartsev and V. Sundström, J. Phys. Chem. A, 2011, 115,
3679.; (b) S. Ye, A. Kathiravan, H. Hayashi, Y. Tong, Y.
Infahsaeng, P. Chabera, T. Pascher, A. P. Yartsev, S. Isoda, H.
Imahori and V. Sundström, J. Phys. Chem. C, 2013, 117, 6066.
22. Q. Wang, JꢀE. Moser, and M. Grätel, J. Phys. Chem. B, 2005, 109,
14945.
23. D. P. Hagberg, T. Edvinsson, T. Marinado, G. Boschloo, A. Hagfeldt
and L. Sun, Chem. Commun., 2006, 2245.
24. W. Kim, T. Tachikawa, T. Majima, C. Li, HꢀJ. Kim and W. Choi,
Energy Environ. Sci.
,
2010,
3 , 1789.
25. A. D. Becke, J. Chem. Phys.
,
1993, 98, 5648.
26. Y. Zhao and D. G. Truhlar. Theor. Chem. Acc., 2008, 120, 215.
27. a) L. Pintilie, I. Vrejoiu, D. Hesse and M. Alexe, J. Appl. Phys. 2008,
104, 114101. b) L. V. Whitney, Phys. Rev. 1936, 50, 1154.
28. a) M. Tomkiewicz, Y. S. Huang and F. H. Pollak, J. Electrochem.
Soc. 1983, 130, 1514.; b) H. C. Zhong, G. Heuss and V. Misra,
IEEE Electron Device Lett., 2000, 21, 593; c) S. Han, S. K. Kim, C.
S. Hwang, ECS Transaction, 2010, 33, 111.
29. a) I. M. Chan, TꢀY. Hsu and F. C. Hong, Appl. Phys. Lett., 2002 81,
1899.; b) K. Nakaoka, J. Ueyama and K. Ogura, J. Electroanal.
Chem., 2004, 571, 93.; c) J. M. McKay and V. E. Henrich, Phys.
Rev. B, 1985, 32, 6764.
30. Y. Mizukoshi, Y. Makise, T. Shuto, J. Hu, A. Tominaga, S. Shironita
and S. Tanabe, Ultrason Sonochem, 2007, 14, 387.
31. G. Kresse and J. Hafner, Phys. Rev. B 1993, 47, 558.ꢀ
32. G. Kresse and J. Furthmueller, Phys. Rev. B 1996, 54, 11169.ꢀ
33. G. Kresse and J. Furthmueller, Comput. Mater. Sci. 1996, 6, 15.
34. a) T. Higashino and H. Imahori, Dalton trans., 2015, 44, 448.; b) A.
S. Hart, B. K. C. Chandra, H. B. Gobeze, L. R. Sequeira and F. D’
Souza, ACS Appl. Mater. Interfaces, 2013, 5, 5314.; c) H. Imahori,
S. Kang, H. Hayashi, M. Haruta, H. Kurata, S. Isoda, S. E. Canton,
Y. Infahsaeng, A. Kathiravan, T. Pascher, P. Chábera, A. P. Yartsev
and V. Sundström, J. Phys. Chem. A, 2011, 115, 3679.; d) S. Ye, A.
Kathiravan, H. Hayashi, Y. Tong, Y. Infahsaeng, P. Chabera, T.
Pascher, A. P. Yartsev, S. Isoda, H. Imahori and V. Sundström, J.
Phys. Chem. C, 2013, 117, 6066.
35. J. E. Kroeze, N. Hirata, S. Koops, Md. K. Nazeeruddin, L. Schmidtꢀ
Mende, M. Grätzel and J. R. Durrant, J. Am. Chem. Soc., 2006, 128,
16376.
36. .L. SchmidtꢀMende, J. E. Kroeze, J. R. Durrant, Md. K. Nazeeruddin
and M. Grätzel, Nano Lett., 2005, 7, 1315.
37. M. Grätzel, Acc. Chem. Res., 2009, 42, 1788.
38. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A.
Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F.
Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F.
Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N.
Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A.
Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M.
Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J.
Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R.
Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G.
Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A.
D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, and D.
J. Fox, Gaussian, Inc., Wallingford CT, 2010.
39. K. Momma and F. Izumi, J. Appl. Crystallogr. 2011, 44, 1272.
40. N. Koumura and K. Hara, Heterocycles, 2013, 83, 275.
8 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 2012

Page 9 of 9
Journal of Materials Chemistry A
View Article Online
DOI: 10.1039/C5TA04991A
185x82mm (144 x 144 DPI)